Mixed venous blood gas sampling is not influenced by the speed of withdrawal in cardiac surgery patients

Objective To determine whether mixed venous blood gas sampling obtained by pulmonary artery catheter (PAC) is influenced by the speed of withdrawal.Design Prospective, observational study.Setting Surgical intensive care unit at a university hospital.Subjects Twenty-five patients in the early postoperative period of cardiac surgery.Measurements and main results After verification of the adequate position of the PAC, measurements of mixed venous blood gas oxygen saturation, oxygen partial pressure (PO₂), carbon dioxide partial pressure (PCO₂), pH and bicarbonates were obtained at two different rates of withdrawal. A slow sampling was taken at a mean speed of 3 ml/min and a fast sampling at 18 ml/min for each patient. The mean difference in venous oxygen saturation between slow and fast samplings was -0.18±1.3%, venous PO₂: –0.2±1.3 mmHg, venous PCO₂: 0.1±0.9 mmHg, venous pH: 0±0.03, venous bicarbonates: 0.03±0.5 mmol/l.Conclusion Using the Bland & Altman method, we showed a satisfactory agreement between slow and fast mixed venous blood gas sampling techniques when measuring PO₂, oxygen saturation, PCO₂, pH and bicarbonates though a PAC.Electronic Supplementary Material Supplementary material is available in the online version of this article at http://dx.doi.org/10.1007/s00134.004.2392-4This study was supported by the Plan de pratique des anesthésiologistes de lInstitut de Cardiologie de Montréal.